Project Overview
Objective
Used parcel-and-person media assets to explore persistent identities across video frames.
Stack
Computer Vision DatasetBoT-SORTByteTrackVideo Test Assets
Delivery highlights
- Explored how BoT-SORT and ByteTrack can associate parcel and person detections across consecutive video frames and assign persistent IDs for visual review.
- The referenced audit root is primarily a dataset and test-asset collection, not a complete runnable tracking application. The demo videos document experimental output, while tracker implementation details and benchmark metrics were not independently confirmed in the reviewed snapshot.